text_alignment 0.12.1 → 0.12.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '097a3da5324e29b872a7927e2725ce82120f328ce1961cb24aaa5e058cf1b265'
-  data.tar.gz: 727bcfbffe4c30535a3b0bd01f4d14cdd1f4ccf64c5f337549da116e484e5e35
+  metadata.gz: 3c2a36fe4cfde7dfb76f554fd4afcae7cb5a03e455887621217f5e5e633b20b3
+  data.tar.gz: f63070c6f423bc15d0fc8c742a21238a104a1b2c1d3fe56ac436effa8ef8eacf
 SHA512:
-  metadata.gz: 0bb88bc2f353bca72b1cc1488063c3ac3d9f3164c3d877bbef8b17876311a99f08143fced8d8ccaa7283cce1b4e6a1ba265435ef71d6c30979da0a5045462290
-  data.tar.gz: ef1a61192ade8cabe0b8960619f3200e1312b3cb6766d4b4dfc4a4685ab8af8397c9647cbc6e50cc6047b5ca4a2daaf7b47e669655616082fa99c06f6d3f1e26
+  metadata.gz: '078a41bc6ab9b16e7747be6a3bb15aff4b23a1161bcea0b653a93f04d673799afcac2109cd1ce8d1a95c99c5c07d36842e3698c9f0997500e653fb4ab939e04a'
+  data.tar.gz: ce44d334779d43b3057317537f615ebf39b8049639d3e50c4e14272c952b76a5df2b060cfd3a15f1d28372c11a795ef8bf43cb04de32b5a78ce2f44433edddfb

data/lib/text_alignment/char_mapping.rb

@@ -93,6 +93,7 @@ class TextAlignment::CharMapping
 
 		@text  = _text
 
+		# sort by the length of the spell-outs is important
 		char_mapping ||= TextAlignment::CHAR_MAPPING.sort{|a, b| b[1].length <=> a[1].length}
 		@mapped_text, offset_mapping = enmap_text(_text, char_mapping)
 		@index_enmap = offset_mapping.to_h
@@ -137,7 +138,7 @@ class TextAlignment::CharMapping
 	def enmap_text(_text, char_mapping, no_ws = false)
 		text = _text.dup
 
-		# To execute the single letter mapping replacement
+		# To perform the single letter mapping replacement
 		char_mapping.each do |one, long|
 			text.gsub!(one, long) if long.length == 1
 		end
@@ -149,12 +150,16 @@ class TextAlignment::CharMapping
 
 			init_next = 0
 			while loc = text.index(long, init_next)
-				rpositions << [loc, long.length, 1]
+				# Huristics to check if the surrounding letters are sufficiently distinguished.
+				if long.length > 3 || ((text[loc - 1, 2] !~ /[a-z][a-z]/) && (text[loc + long.length - 1, 2] !~ /[a-z][a-z]/))
+				# if true
+					rpositions << [loc, long.length, 1]
+
+					# a workaround to avoid messing-up due to embedding
+					text[loc, long.length] = one * long.length
+				end
 				init_next = loc + long.length
 			end
-
-			# a workaround to avoid messing-up due to embedding
-			text.gsub!(long, one * long.length)
 		end
 
 		# To get the replacement positions, (position, old_length, new_length), for consecutive whitespaces
@@ -182,6 +187,7 @@ class TextAlignment::CharMapping
 
 		# To execute the long letter mapping
 		char_mapping.each do |one, long|
+			next unless text =~ /#{one}/
 			text.gsub!(one * long.length, one) if long.length > 1
 		end
 

data/lib/text_alignment/mixed_alignment.rb

@@ -17,7 +17,7 @@ class TextAlignment::MixedAlignment
 	attr_reader :similarity
 	attr_reader :str1_match_initial, :str1_match_final, :str2_match_initial, :str2_match_final
 
-	def initialize(_str1, _str2, _mappings = nil)
+	def initialize(_str1, _str2, mappings = nil)
 		raise ArgumentError, "nil string" if _str1.nil? || _str2.nil?
 
 		mappings ||= TextAlignment::CHAR_MAPPING

data/lib/text_alignment/version.rb

@@ -1,3 +1,3 @@
 class TextAlignment
-	VERSION = '0.12.1'
+	VERSION = '0.12.2'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: text_alignment
 version: !ruby/object:Gem::Version
-  version: 0.12.1
+  version: 0.12.2
 platform: ruby
 authors:
 - Jin-Dong Kim
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-12-26 00:00:00.000000000 Z
+date: 2024-01-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-dictionary